Fiddler AI
Fiddler's mission is to build trust into AI by providing visibility, evaluation, observability, security, policy enforcement, and governance for AI systems. It positions its platform as the AI Control Plane for first-party and third-party agents from creation through production.
At a Glance
- Large enterprises and Fortune 500 companies
- Software and AI engineering teams
- Data science and ML platform teams
- DevOps, security, risk, and compliance teams
- +2 more
AI Tools by Fiddler AI
(1)
Fiddler AI
AI Agent Observability and Governance
Discussions
No discussions yet
Be the first to start a discussion about Fiddler AI
Latest News
Fiddler raised $30M Series C led by RPS Ventures, bringing total funding to $100M and advancing its AI Control Plane for agents.
Fiddler AI Observability and Security selected as an AWS Pattern Partner.
Fiddler enhanced agentic-AI security with NVIDIA Enterprise AI Factory integration.
Fiddler Guardrails became native to NVIDIA NeMo Guardrails.
Products & Services
A centralized system of trust for AI agents, providing standardized telemetry, evaluation, continuous monitoring, enforceable policy, and auditable governance across the AI lifecycle.
Observability for production agentic applications, including visibility into agent actions, traces, tool calls, tokens, cost, latency, and outcomes.
Monitoring, explainability, metrics, dashboards, alerts, drift, bias, data integrity, and root-cause analysis for predictive ML models and LLM applications.
Continuous pre-deployment and production evaluation that closes the loop between testing, monitoring, failure analysis, and improvement of agents.
Market Position
Fiddler competes with AI/ML observability and evaluation platforms such as Arize, while positioning itself beyond monitoring through unified observability, low-latency runtime guardrails, policy enforcement, and governance across both agent creation and production. Its differentiation is the Control Plane framing and the use of in-environment Centor Models to reduce external-LLM evaluation cost and latency.
Leadership
Founders
Krishna Gade
Founder and CEO; prior experience at Facebook, Twitter, Pinterest, and Microsoft.
Amit Paka
Founder and COO; prior experience at Samsung, PayPal, and Microsoft; attended UC Berkeley.
Manoj Cheenath
Co-founder and Chief Architect at the company's 2021 Series B announcement.
Executive Team
Krishna Gade
Founder and CEO
Previously associated with Facebook, Twitter, Pinterest, and Microsoft.
Amit Paka
Founder and COO
Previously associated with Samsung, PayPal, and Microsoft; UC Berkeley.
Board of Directors
Founding Story
The founders experienced the early pain of opaque machine-learning models firsthand and started Fiddler to solve it, initially through explainability. The company expanded into observability to track how models change over time, and later into governance and control as autonomous agents began making consequential enterprise decisions.
Business Model
Revenue Model
B2B SaaS and enterprise software sold through tiered plans, usage-based trace pricing for the Developer plan, and customized enterprise contracts. Deployment can be SaaS, VPC, or on-premises.
Pricing Tiers
Real-time guardrails for hallucinations, toxicity, PII/PHI, prompt injection, and jailbreaks; under 80ms latency; powered by Fiddler Centor Models.
Adds unified observability, tests and experiments, custom evaluators/bring-your-own-judge, visualization-driven insights, RBAC, SSO, and SaaS deployment.
Adds enterprise guardrails and scale, SaaS/VPC/on-premises deployment, white-glove support, named Customer Success Manager, and customized onboarding.
Target Markets
- Large enterprises and Fortune 500 companies
- Software and AI engineering teams
- Data science and ML platform teams
- DevOps, security, risk, and compliance teams
- Financial services, advertising, healthcare, insurance, retail, and technology
- Government and defense organizations
- Enterprise AI-agent observability and governance
- Production monitoring of LLM, RAG, and generative-AI applications
- Predictive ML model performance management
- Pre-deployment and continuous evaluation of agents
- Runtime AI safety, security, compliance, and policy enforcement
- Responsible AI, model drift, bias, and data-quality management
- Nielsen
- Integral Ad Science
- U.S. Department of Energy
- Qualtrics